feat(agent): confine agent file/shell tools to a selectable workspace (#3665)

* feat(agent): workspace confinement via context-local binding + get_workspace tool

Bind the per-turn workspace once in execute_tool_block; the shared path
resolvers (_resolve_tool_path / _resolve_search_root) and the subprocess cwd
helper (agent_cwd) read it, so file tools + bash/python are confined centrally
and a new tool that uses the shared helpers cannot accidentally bypass it.

Adds the admin-gated /api/workspace/browse picker, a workspace pill + directory
modal (reusing existing modal/button CSS), the /workspace slash command, and a
get_workspace tool (replaces a system-prompt block). Confinement is OS-agnostic
(realpath/normcase/commonpath) and docker-safe (container paths, no host
assumptions). Reopens #2023.

* ux(workspace): clarify workspace is not a sandbox

Picker modal note + pill tooltip + get_workspace tool/output wording now state
plainly: read_file/write_file/edit_file/grep/glob/ls are confined to the folder,
but bash/python only start there (cwd) and are not sandboxed. Modal note reuses
the existing .muted class.

* fix(agent): treat an active workspace as file-work intent

A vague low-signal message (e.g. "look at the local project") matches no
domain keywords, so tool retrieval is skipped and only always-available tools
are offered — leaving the agent with no file access even though a workspace is
set. When a workspace is active, include the file/code tools (incl.
get_workspace) on low-signal turns so the agent can act on the folder.

Also requires the tool index (ChromaDB) to be reachable for normal retrieval;
that is an environment dependency, not part of this change.

* ux(workspace): hide pill + overflow entry in chat mode

Workspace only scopes the agent's file/shell tools, so the pill and the
overflow 'Workspace' entry are agent-only now — hidden in chat mode like the
bash toggle. Mode read from the DOM in syncWorkspaceIndicator; applyMode() is
called from the agent/chat setMode handler.

* prompt(tools): steer bash/python to defer to the dedicated file tools

bash/python schema descriptions (what native-tool-calling models read) were
bare and gave no steer, so models would do file ops via the shell (e.g. writing
SVG/HTML, which then dumps raw markup into the tool preview). Tell bash/python
in the schema + tool-index + prompt section to prefer read_file/write_file/
edit_file/grep/glob/ls and only be used for what those do not cover.

* prompt(tools): keep bash/python deferral generic (no hardcoded tool names)

Reference 'a dedicated tool' rather than listing read_file/write_file/grep/etc.
by name, so the guidance does not go stale if those tools are renamed.

* style(workspace): drop em-dashes from added code comments/strings

* ux(workspace): terser non-sandbox note in picker (no tool-name list)

* ux(workspace): mirror terse non-sandbox wording in pill tooltip

* chore: untrack local venv symlink (run-only, not part of the feature)

* prompt(workspace): keep get_workspace text generic (no hardcoded tool names)

* fix(agent): low-signal + workspace surfaces only read-only file tools

Intersect the files tool group with PLAN_MODE_READONLY_TOOLS so a vague message
in a workspace exposes read_file/grep/glob/ls/get_workspace for exploration, but
not write_file/edit_file/bash/python -- those wait for a request that actually
calls for them (RAG retrieval still adds them on a real ask).

* feat(workspace): cap browse listing at 500 dirs with a truncated hint

Mirror the filesystem_tools._CODENAV_MAX_HITS pattern with a module-local
_MAX_BROWSE_DIRS so a directory with thousands of children does not dump every
row into the picker; the response carries a truncated flag and the modal tells
the user to type a path to jump in.

* chore: untrack local venv symlink (run-only artifact)

* fix(workspace): vet the workspace root against the sensitive-path deny list at bind time

The in-workspace resolver deny-lists sensitive paths inside the workspace,
but the empty-path search root is the workspace itself, so a workspace of
~/.ssh could be listed via ls with no path. vet_workspace() (public, in
tool_execution next to the resolvers) rejects non-directories and sensitive
roots before the path is ever bound; chat_routes uses it instead of its
inline isdir check.

* fix(workspace): reject filesystem roots and stop showing rejected workspaces as active

Review findings from #3665:

P2: vet_workspace accepted / (and would accept drive/UNC roots), which makes
every absolute path 'inside' the workspace and collapses confinement into
host-wide file access. A root is its own dirname, so reject when
dirname(resolved) == resolved; the browse response now carries a selectable
flag and the picker disables 'Use this folder' on unselectable dirs.

P3: /workspace set stored any string client-side and the chat route silently
dropped rejected values, so the pill could claim a confinement that was not
in effect. New admin-gated /api/workspace/vet validates manual paths before
they persist (canonical path returned), and when a posted workspace is
rejected at send time the stream emits workspace_rejected so the client
clears the stored value and toasts instead of continuing silently.

* fix(workspace): check caller privilege before vetting the posted workspace

Review finding: /api/chat_stream called vet_workspace() on the posted value
for every caller and emitted workspace_rejected on failure, so a non-admin
who can chat but cannot use file/shell tools could distinguish existing
directories from missing/file/sensitive/root paths by whether the event
appeared. The resolution now lives in _resolve_request_workspace, which
drops the submitted value uniformly for non-admin callers, with no vetting
and no event, before the path ever touches the filesystem. Admin and
single-user behavior is unchanged. Test pins that valid and invalid paths
are indistinguishable for a non-admin and that vet_workspace is never
invoked for them.

src/agent_loop.py

@@ -272,7 +272,7 @@ _DOMAIN_TOOL_MAP = {
     "notes_calendar_tasks": {"manage_notes", "manage_calendar", "manage_tasks"},
     "ui": {"ui_control"},
     "sessions": {"create_session", "list_sessions", "manage_session", "send_to_session", "search_chats"},
-    "files": {"bash", "python", "read_file", "write_file", "edit_file", "grep", "glob", "ls"},
+    "files": {"bash", "python", "read_file", "write_file", "edit_file", "grep", "glob", "ls", "get_workspace"},
     "settings": {"manage_settings", "manage_endpoints", "manage_mcp", "manage_webhooks", "manage_tokens", "app_api"},
 }
 
@@ -309,6 +309,7 @@ NEVER pipe multi-line Python through `python -c "..."` — shell quoting eats re
 <python code>
 ```
 Execute Python code. Use for computation, data processing, scripting. NOT for writing code for the user (use create_document for that). Same sandbox limits as bash — no TTY, no GUI, no `input()`; for anything the user should interact with, generate a single HTML file with inline JS instead.
+Prefer a dedicated tool whenever one fits the job (reading, searching, or writing files); use python only for computation/processing no dedicated tool covers - not for reading or writing files.
 Do NOT use Python/requests for web lookup/search/latest/current requests when `web_search` or `web_fetch` is available.""",
 
     "web_search": """\
@@ -347,6 +348,11 @@ Write content to a file. First line is the path, rest is the content.""",
 ```
 Edit an EXISTING file by exact string replacement. PREFER this over bash (sed/echo/redirects) for changing files — it shows a before/after diff. `old_string` must match the file exactly and be unique unless `replace_all` is true. Use write_file to create a new file.""",
 
+    "get_workspace": """\
+```get_workspace
+```
+Return the absolute path of the active workspace folder. File tools are CONFINED to it (paths can be RELATIVE to it); the shell starts there (cwd) but is NOT sandboxed. Call this first when the user says "the project"/"the code"/"this folder" without a path, instead of asking them. No arguments.""",
+
     "create_document": """\
 ```create_document
 <title>
@@ -1726,6 +1732,7 @@ async def stream_agent_loop(
     plan_mode: bool = False,
     approved_plan: Optional[str] = None,
     tool_policy: Optional[ToolPolicy] = None,
+    workspace: Optional[str] = None,
     _is_teacher_run: bool = False,
 ) -> AsyncGenerator[str, None]:
     """Streaming agent loop generator.
@@ -1795,7 +1802,17 @@ async def stream_agent_loop(
     if not guide_only and not _relevant_tools and bool(_intent.get("low_signal")):
         from src.tool_index import ALWAYS_AVAILABLE
         _relevant_tools = set(ALWAYS_AVAILABLE)
-        logger.info("[tool-rag] Low-signal agent message; skipping retrieval and using always-available tools only")
+        if workspace:
+            # An active workspace IS the file-work signal: a vague "look at the
+            # project" means explore this folder. Surface only the READ-ONLY file
+            # tools (intersection with the plan-mode read-only allowlist) so the
+            # agent can investigate; write/shell tools stay out until the request
+            # actually calls for them (RAG retrieval adds those on a real ask).
+            from src.tool_security import PLAN_MODE_READONLY_TOOLS
+            _relevant_tools |= (_DOMAIN_TOOL_MAP["files"] & PLAN_MODE_READONLY_TOOLS)
+            logger.info("[tool-rag] Low-signal but workspace active; including read-only file tools")
+        else:
+            logger.info("[tool-rag] Low-signal agent message; skipping retrieval and using always-available tools only")
     if not guide_only and not _relevant_tools:
         try:
             from src.tool_index import get_tool_index, ALWAYS_AVAILABLE
@@ -2644,6 +2661,7 @@ async def stream_agent_loop(
                             tool_policy=tool_policy,
                             owner=owner,
                             progress_cb=_push_progress,
+                            workspace=workspace,
                         )
                     finally:
                         # Sentinel so the drainer knows to stop.